Assess Dataset Annotation Coverage — assess_annotation

Checks what proportion of derived variables have derivation text documented. Returns one evidence row per dataset summarising annotation coverage, with indicator_id = "U-003".

Usage

assess_annotation_coverage(metadata, ctx, config = NULL)

Arguments

metadata: A data frame with columns dataset, variable, origin, and derivation.
ctx: A run context created by r4subcore::r4sub_run_context().
config: A configuration list from usability_config_default(). If NULL the default configuration is used.

Value

A validated evidence tibble (see r4subcore::as_evidence()).

Examples

ctx <- r4subcore::r4sub_run_context(study_id = "STUDY01")
#> ℹ Run context created: "R4S-20260316173057-wl4dieex"
meta <- data.frame(
  dataset    = c("ADSL", "ADSL", "ADSL"),
  variable   = c("AGE", "SEX", "RACE"),
  origin     = c("Derived", "CRF", "Derived"),
  derivation = c("Derived from BRTHDTC", NA, NA),
  stringsAsFactors = FALSE
)
ev <- assess_annotation_coverage(meta, ctx)
#> ✔ Evidence table created: 1 row
ev$metric_value
#> [1] 0.5